{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IJWCJCPDBTFID3NV3NGM7HQTGW","short_pith_number":"pith:IJWCJCPD","schema_version":"1.0","canonical_sha256":"426c2489e30cca81edb5db4ccf9e13358f1eda4575934fee9a4db9059e2d1c81","source":{"kind":"arxiv","id":"2605.13297","version":1},"attestation_state":"computed","paper":{"title":"PaMM: Periodic Motif Memory for Atomistic Models with an Explicit Local-Structure Interface","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Explicit pair and triplet motif lookup tables improve energy and force accuracy in periodic atomistic models at intermediate training steps.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ryan Dong","submitted_at":"2026-05-13T10:11:27Z","abstract_excerpt":"Periodic crystals repeatedly instantiate similar local coordination motifs across translated cells and chemically related structures, but current equivariant atomistic models usually encode these patterns only implicitly in dense edge features. We introduce PaMM, a periodic motif memory that augments the UMA eSCN-MD edge encoder with explicit pair and triplet lookup features. Pair motifs are keyed by $(Z_j, Z_i, b_r)$ and triplet motifs by $(Z_j, Z_i, Z_k, b_\\theta)$, hashed into fixed-size tables and fused with the baseline edge representation through lightweight gate-only and affine-equipped"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":true,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.13297","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T10:11:27Z","cross_cats_sorted":[],"title_canon_sha256":"e83ac2c615c2dc95206747807d9fe50a3ddf0775872fd6f7a5913ad4438cca4a","abstract_canon_sha256":"2de67347fa5fd51b8be8bfac5f2a76a7e541a297e5d5d5e79ed6546ee20d3bbe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:49.073379Z","signature_b64":"4qDh9u7Eh9o0DDTJQXF6tbFjupCOG5L/SjWkJNXZMZ2hirk/c3UCqKtplf6msuimiPMFQhdGHH811L7gq4KFAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"426c2489e30cca81edb5db4ccf9e13358f1eda4575934fee9a4db9059e2d1c81","last_reissued_at":"2026-05-18T02:44:49.072971Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:49.072971Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PaMM: Periodic Motif Memory for Atomistic Models with an Explicit Local-Structure Interface","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Explicit pair and triplet motif lookup tables improve energy and force accuracy in periodic atomistic models at intermediate training steps.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ryan Dong","submitted_at":"2026-05-13T10:11:27Z","abstract_excerpt":"Periodic crystals repeatedly instantiate similar local coordination motifs across translated cells and chemically related structures, but current equivariant atomistic models usually encode these patterns only implicitly in dense edge features. We introduce PaMM, a periodic motif memory that augments the UMA eSCN-MD edge encoder with explicit pair and triplet lookup features. Pair motifs are keyed by $(Z_j, Z_i, b_r)$ and triplet motifs by $(Z_j, Z_i, Z_k, b_\\theta)$, hashed into fixed-size tables and fused with the baseline edge representation through lightweight gate-only and affine-equipped"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"in the studied UMA-S + OMAT regime, explicit pair/triplet motif memory is a useful inductive bias for periodic atomistic modeling.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the observed gains at the 10k- and 20k-step checkpoints are driven by the structured motif organization rather than incidental capacity or optimization effects, and that the same pattern would persist at full convergence or on other datasets.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"PaMM augments an equivariant atomistic model with explicit pair and triplet motif memory tables, producing modest gains in energy and force MAE on OMAT benchmarks at fixed training budget.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Explicit pair and triplet motif lookup tables improve energy and force accuracy in periodic atomistic models at intermediate training steps.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"025e6c8f00c6919b0e368bbf21713b97c8e42afd231a1a8f2c427face3b161a6"},"source":{"id":"2605.13297","kind":"arxiv","version":1},"verdict":{"id":"03136ba0-f472-4b35-be0e-eaf89db90647","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T19:28:35.463373Z","strongest_claim":"in the studied UMA-S + OMAT regime, explicit pair/triplet motif memory is a useful inductive bias for periodic atomistic modeling.","one_line_summary":"PaMM augments an equivariant atomistic model with explicit pair and triplet motif memory tables, producing modest gains in energy and force MAE on OMAT benchmarks at fixed training budget.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the observed gains at the 10k- and 20k-step checkpoints are driven by the structured motif organization rather than incidental capacity or optimization effects, and that the same pattern would persist at full convergence or on other datasets.","pith_extraction_headline":"Explicit pair and triplet motif lookup tables improve energy and force accuracy in periodic atomistic models at intermediate training steps."},"references":{"count":10,"sample":[{"doi":"","year":2025,"title":"UMA: A Family of Universal Models for Atoms , author=. 2025 , eprint=","work_id":"1fd6d722-4575-4e4b-a7cc-90f6aa8c2813","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2026,"title":"Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models , author=. 2026 , eprint=","work_id":"e369fb3d-e109-4fb1-924f-c4b6fdd809e0","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Nature Communications , volume=","work_id":"52bf5550-8992-43a8-9cde-67e2b6a68cca","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields , author=. 2022 , eprint=","work_id":"7d4e0f06-27cc-459f-a86e-618a9627556c","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representations , author=. 2023 , eprint=","work_id":"4ae132e1-8fd9-47ce-9f9b-9cdc453ce282","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":10,"snapshot_sha256":"9e8c41683c862811ac95fe0804eae28c6418f57ce9b80f52c0ff464a3ed333fe","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.13297","created_at":"2026-05-18T02:44:49.073029+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.13297v1","created_at":"2026-05-18T02:44:49.073029+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13297","created_at":"2026-05-18T02:44:49.073029+00:00"},{"alias_kind":"pith_short_12","alias_value":"IJWCJCPDBTFI","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"IJWCJCPDBTFID3NV","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"IJWCJCPD","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IJWCJCPDBTFID3NV3NGM7HQTGW","json":"https://pith.science/pith/IJWCJCPDBTFID3NV3NGM7HQTGW.json","graph_json":"https://pith.science/api/pith-number/IJWCJCPDBTFID3NV3NGM7HQTGW/graph.json","events_json":"https://pith.science/api/pith-number/IJWCJCPDBTFID3NV3NGM7HQTGW/events.json","paper":"https://pith.science/paper/IJWCJCPD"},"agent_actions":{"view_html":"https://pith.science/pith/IJWCJCPDBTFID3NV3NGM7HQTGW","download_json":"https://pith.science/pith/IJWCJCPDBTFID3NV3NGM7HQTGW.json","view_paper":"https://pith.science/paper/IJWCJCPD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.13297&json=true","fetch_graph":"https://pith.science/api/pith-number/IJWCJCPDBTFID3NV3NGM7HQTGW/graph.json","fetch_events":"https://pith.science/api/pith-number/IJWCJCPDBTFID3NV3NGM7HQTGW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IJWCJCPDBTFID3NV3NGM7HQTGW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IJWCJCPDBTFID3NV3NGM7HQTGW/action/storage_attestation","attest_author":"https://pith.science/pith/IJWCJCPDBTFID3NV3NGM7HQTGW/action/author_attestation","sign_citation":"https://pith.science/pith/IJWCJCPDBTFID3NV3NGM7HQTGW/action/citation_signature","submit_replication":"https://pith.science/pith/IJWCJCPDBTFID3NV3NGM7HQTGW/action/replication_record"}},"created_at":"2026-05-18T02:44:49.073029+00:00","updated_at":"2026-05-18T02:44:49.073029+00:00"}